Rewrite the letter that got no calls
Lesson 6 of 6 · Mail that gets a phone call
1 · Learn the move · Refinement loop
After a flat drop the reflex is to blame the letter and ask for a better one. The model will write a punchier one and imply it will do better, which it cannot know. The honest loop has three turns. First, the critique, fenced to what you pasted: the letter, how the list was built, what went out, what came back. It separates what the words could have caused from what the list, the offer and the timing could have caused, and it says plainly that it cannot judge whether your offers were low, because it cannot price. Second, a rewrite that fixes only what the critique found in the words. Third, the rewrite critiqued against the same list, then improved. What comes out is a letter and a test you can read with your own counts, never a promised lift and never a 'normal' response rate borrowed from nowhere.
This letter went to 700 owners and got 3 calls. Before rewriting anything, critique it: separate what the wording could have caused from what the list, the offer or the timing could have caused. Do not judge my offer amounts. Then rewrite, critique your rewrite, improve it, and give me a split test to run. No predicted response rate.
2 · Your turn. You write the prompt
A 700-piece blind offer drop in Sayle County got three calls and one of them was angry. The same letter did better last spring somewhere else. You want to know whether to rewrite it or look elsewhere first, and you want the rewrite either way. Write the prompt.
Write the prompt for this scenario as given, or swap in a real one from your own work. The score works the same either way.
Remember: the AI sees only your prompt, not this page. If the situation isn't in your prompt, it doesn't exist.
Optional. These shape the output when you run your prompt below, not your score.